Dynamic adaptive mechanism for construction investment decision-making under multi-factor interactions: risk-oriented modelling framework with pathways to digital twin

Abstract

Construction project investment decision-making faces challenges from dynamic multi-factor interactions, where traditional static models fail to address policy-market-technology coupling effects and the transmission of decision-makers? cognitive biases. This study proposes a risk-oriented dynamic adaptive mechanism by integrating systems theory and complexity science to construct a multi-level networked factor system, revealing synergistic-constraining mechanisms among economic, technical, social, and environmental factors. Methodologically, it innovatively couples System Dynamics with real-time data streams and embeds neuro-economics agent models to establish a dynamic feedback analysis framework. Empirical validation via a digital twin platform demonstrates significant efficacy: an 85% reduction in conflict risks through policy digitisation in the Suzhou Low-altitude Project; a 90% improvement in response efficiency (from 10?minutes to 1?minute) for smart dispatching in the Nangang River Project; and seepage control verticality error maintained below 0.1% in the Huaihe River Project. The proposed multi-factor interaction analysis framework grounded in systems theory and complexity science provides a scalable decision-support paradigm for complex investment environments, bridging critical academic gaps in quantitative tools for interaction analysis and dynamic adaptation mechanisms.